Automatic Segmentation of Epidermis and Hair Follicles in Optical Coherence Tomography Images of Normal Skin by Convolutional Neural Networks

Segmentation of 3D OCT Images of Human Skin Using Neural Networks with U-Net Architecture

The aim of the study is a comparative analysis of algorithms for segmentation of three-dimensional OCT images of human skin using neural networks based on U-Net architecture when training the model on two-dimensional and three-dimensional data.

Materials and Methods

Two U-Net-based network architectures for segmentation of 3D OCT skin images are proposed in this work, in which 2D and 3D blocks of 3D images serve as input data. Training was performed on thick skin OCT images acquired from 7 healthy volunteers. For training, the OCT images were semi-automatically segmented by experts in OCT and dermatology. The Sørensen-Dice coefficient, which was calculated from the segmentation results of images that did not participate in the training of the networks, was used to assess the quality of segmentation. Additional testing of the networks' capabilities in determining skin layer thicknesses was performed on an independent dataset from 8 healthy volunteers.

Results

In evaluating the segmentation quality, the values of the Sørensen-Dice coefficient for the upper stratum corneum, ordered stratum corneum, epidermal cellular layer, and dermis were 0.90, 0.94, 0.89, and 0.99, respectively, for training on two-dimensional data and 0.89, 0.94, 0.87, and 0.98 for training on three-dimensional data. The values obtained for the dermis are in good agreement with the results of other works using networks based on the U-Net architecture. The thicknesses of the ordered stratum corneum and epidermal cellular layer were 153±24 and 137±17 μm, respectively, when the network was trained on two-dimensional data and 163±19 and 137±20 μm when trained on three-dimensional data.

Conclusion

Neural networks based on U-Net architecture allow segmentation of skin layers on OCT images with high accuracy, which makes these networks promising for obtaining valuable diagnostic information in dermatology and cosmetology, e.g., for estimating the thickness of skin layers.

Harmonizing Unets: Attention Fusion module in cascaded-Unets for low-quality OCT image fluid segmentation

Optical coherence tomography (OCT) is widely used for its high resolution. Accurate OCT image segmentation can significantly improve the diagnosis and treatment of retinal diseases such as Diabetic Macular Edema (DME). However, in resource-limited regions, portable devices with low-quality output are more frequently used, severely affecting the performance of segmentation. To address this issue, we propose a novel methodology in this paper, including a dedicated pre-processing pipeline and an end-to-end double U-shaped cascaded architecture, H-Unets. In addition, an Adaptive Attention Fusion (AAF) module is elaborately designed to improve the segmentation performance of H-Unets. To demonstrate the effectiveness of our method, we conduct a bunch of ablation and comparative studies on three open-source datasets. The experimental results show the validity of the pre-processing pipeline and H-Unets, achieving the highest Dice score of 90.60%±0.87% among popular methods in a relatively small model size.

LS-Net: lightweight segmentation network for dermatological epidermal segmentation in optical coherence tomography imaging

Optical coherence tomography (OCT) can be an important tool for non-invasive dermatological evaluation, providing useful data on epidermal integrity for diagnosing skin diseases. Despite its benefits, OCT's utility is limited by the challenges of accurate, fast epidermal segmentation due to the skin morphological diversity. To address this, we introduce a lightweight segmentation network (LS-Net), a novel deep learning model that combines the robust local feature extraction abilities of Convolution Neural Network and the long-term information processing capabilities of Vision Transformer. LS-Net has a depth-wise convolutional transformer for enhanced spatial contextualization and a squeeze-and-excitation block for feature recalibration, ensuring precise segmentation while maintaining computational efficiency. Our network outperforms existing methods, demonstrating high segmentation accuracy (mean Dice: 0.9624 and mean IoU: 0.9468) with significantly reduced computational demands (floating point operations: 1.131 G). We further validate LS-Net on our acquired dataset, showing its effectiveness in various skin sites (e.g., face, palm) under realistic clinical conditions. This model promises to enhance the diagnostic capabilities of OCT, making it a valuable tool for dermatological practice.

Rapid measurement of epidermal thickness in OCT images of skin

Epidermal thickness (ET) changes are associated with several skin diseases. To measure ET, segmentation of optical coherence tomography (OCT) images is essential; manual segmentation is very time-consuming and requires training and some understanding of how to interpret OCT images. Fast results are important in order to analyze ET over different regions of skin in rapid succession to complete a clinical examination and enable the physician to discuss results with the patient in real time. The well-known CNN-graph search (CNN-GS) methodology delivers highly accurate results, but at a high computational cost. Our objective was to build a computational core, based on CNN-GS, able to accurately segment OCT skin images in real time. We accomplished this by fine-tuning the hyperparameters, testing a range of speed-up algorithms including pruning and quantization, designing a novel pixel-skipping process, and implementing the final product with efficient use of core and threads on a multicore central processing unit (CPU). We name this product CNN-GS-skin. The method identifies two defined boundaries on OCT skin images in order to measure ET. We applied CNN-GS-skin to OCT skin images, taken from various body sites of 63 healthy individuals. Compared with CNN-GS, our described method reduced computation time by 130 [Formula: see text] with minimal reduction in ET determination accuracy (from 96.38 to 94.67%).

Integrated convolutional neural network for skin cancer classification with hair and noise restoration

Background/aim

Skin lesions are commonly diagnosed and classified using dermoscopic images. There are many artifacts visible in dermoscopic images, including hair strands, noise, bubbles, blood vessels, poor illumination, and moles. These artifacts can obscure crucial information about lesions, which limits the ability to diagnose lesions automatically. This study investigated how hair and noise artifacts in lesion images affect classifier performance and how they can be removed to improve diagnostic accuracy.

Materials and methods

A synthetic dataset created using hair simulation and noise simulation was used in conjunction with the HAM10000 benchmark dataset. Moreover, integrated convolutional neural networks (CNNs) were proposed for removing hair artifacts using hair inpainting and classification of refined dehaired images, called integrated hair removal (IHR), and for removing noise artifacts using nonlocal mean denoising and classification of refined denoised images, called integrated noise removal (INR).

Results

Five deep learning models were used for the classification: ResNet50, DenseNet121, ResNet152, VGG16, and VGG19. The proposed IHR-DenseNet121, IHR-ResNet50, and IHR-ResNet152 achieved 2.3%, 1.78%, and 1.89% higher accuracy than DenseNet121, ResNet50, and ResNet152, respectively, in removing hairs. The proposed INR-DenseNet121, INR-ResNet50, and INR-VGG19 achieved 1.41%, 2.39%, and 18.4% higher accuracy than DenseNet121, ResNet50, and VGG19, respectively, in removing noise.

Conclusion

A significant proportion of pixels within lesion areas are influenced by hair and noise, resulting in reduced classification accuracy. However, the proposed CNNs based on IHR and INR exhibit notably improved performance when restoring pixels affected by hair and noise. The performance outcomes of this proposed approach surpass those of existing methods.

Semantic modeling of cell damage prediction: a machine learning approach at human-level performance in dermatology

Machine learning is transforming the field of histopathology. Especially in classification related tasks, there have been many successful applications of deep learning already. Yet, in tasks that rely on regression and many niche applications, the domain lacks cohesive procedures that are adapted to the learning processes of neural networks. In this work, we investigate cell damage in whole slide images of the epidermis. A common way for pathologists to annotate a score, characterizing the degree of damage for these samples, is the ratio between healthy and unhealthy nuclei. The annotation procedure of these scores, however, is expensive and prone to be noisy among pathologists. We propose a new measure of damage, that is the total area of damage, relative to the total area of the epidermis. In this work, we present results of regression and segmentation models, predicting both scores on a curated and public dataset. We have acquired the dataset in collaborative efforts with medical professionals. Our study resulted in a comprehensive evaluation of the proposed damage metrics in the epidermis, with recommendations, emphasizing practical relevance for real world applications.

Radiation-induced alterations in multi-layered, in-vitro skin models detected by optical coherence tomography and histological methods

BACKGROUND

Inflammatory skin reactions and skin alterations are still a potential side effect in radiation therapy (RT), which also need attention for patients' health care.

METHOD

In a pre-clinical study we consider alterations in irradiated in-vitro skin models of epidermal and dermal layers. Typical dose regimes in radiation therapy are applied for irradiation. For non-invasive imaging and characterization optical coherence tomography (OCT) is used. Histological staining method is additionally applied for comparison and discussion.

RESULTS

Structural features, such as keratinization, modifications in epidermal cell layer thickness and disorder in the layering-as indications for reactions to ionizing radiation and aging-could be observed by means of OCT and confirmed by histology. We were able to recognize known RT induced changes such as hyper-keratosis, acantholysis, and epidermal hyperplasia as well as disruption and/or demarcation of the dermo-epidermal junction.

CONCLUSION

The results may pave the way for OCT to be considered as a possible adjunctive tool to detect and monitor early skin inflammation and side effects of radiotherapy, thus supporting patient healthcare in the future.

Mouse model of selective cryolipolysis

BACKGROUND

Cryolipolysis is a noninvasive method of destroying adipocytes using controlled cooling, thereby enabling localized and targeted fat reduction. Due to their greater vulnerability to cold injury, adipocytes are selectively targeted, while other cell types are spared.

OBJECTIVES

This study aims to develop a mouse model of cryolipolysis to offer a reliable and convenient alternative to human models, providing a methodology to validate clinical hypotheses in-depth with relative ease, low cost, and efficiency. This further facilitates comprehensive studies of the molecular mechanisms involved in cryolipolysis.

MATERIALS AND METHODS

Mice (C57BL/6J) were placed under general anesthesia and were treated using our custom, miniaturized cryolipolysis system. A thermoelectric cooling probe was applied to the inguinal (ING) area for either a cold exposure of -10°C, or for a room temperature exposure for 10 minutes. The thickness of the subcutaneous fat of the mice was quantified using an optical coherence tomography (OCT) imaging system before and after the treatment. Histological analyses were performed before and after cryolipolysis at multiple time points.

RESULTS

OCT analysis showed that mice that underwent cold cryolipolysis treatment induced a significantly greater reduction of subcutaneous fat thickness 1 month after treatment than the control mice. The mice that received cold treatment had no skin injuries. The selective damage of adipocytes stimulated cold panniculitis that was characterized histologically by infiltration of immune cells 2 and 3 days after treatment.

CONCLUSION

This study shows that cryolipolysis performed in mice yields reproducible and measurable subcutaneous fat reduction, consistent with previous studies conducted in humans and pigs. Future studies can utilize the model of selective cryolipolysis developed by our group to further elucidate the cellular and molecular mechanisms of fat cell loss and improve clinical outcomes in humans.

Optical coherence tomography evaluation of vaginal epithelial thickness during CO2 laser treatment: A pilot study

Genitourinary syndrome of menopause (GSM) negatively affects more than half of postmenopausal women. Energy-based therapy has been explored as a minimally invasive treatment for GSM; however, its mechanism of action and efficacy is controversial. Here, we report on a pilot imaging study conducted on a small group of menopause patients undergoing laser treatment. Intravaginal optical coherence tomography (OCT) endoscope was used to quantitatively monitor the changes in the vaginal epithelial thickness (VET) during fractional-pixel CO2 laser treatment. Eleven patients with natural menopause and one surgically induced menopause patient were recruited in this clinical study. Following the laser treatment, 6 out of 11 natural menopause patient showed increase in both proximal and distal VET, while two natural menopause patient showed increase in VET in only one side of vaginal tract. Furthermore, the patient group that showed increased VET had thinner baseline VET compared to the patients that showed decrease in VET after laser treatment. These results demonstrate the potential utility of intravaginal OCT endoscope in evaluating the vaginal tissue integrity and tailoring vaginal laser treatment on a per-person basis, with the potential to monitor other treatment procedures.

Computer-Aided Diagnosis Methods for High-Frequency Ultrasound Data Analysis: A Review

Over the last few decades, computer-aided diagnosis systems have become a part of clinical practice. They have the potential to assist clinicians in daily diagnostic tasks. The image processing techniques are fast, repeatable, and robust, which helps physicians to detect, classify, segment, and measure various structures. The recent rapid development of computer methods for high-frequency ultrasound image analysis opens up new diagnostic paths in dermatology, allergology, cosmetology, and aesthetic medicine. This paper, being the first in this area, presents a research overview of high-frequency ultrasound image processing techniques, which have the potential to be a part of computer-aided diagnosis systems. The reviewed methods are categorized concerning the application, utilized ultrasound device, and image data-processing type. We present the bridge between diagnostic needs and already developed solutions and discuss their limitations and future directions in high-frequency ultrasound image analysis. A search was conducted of the technical literature from 2005 to September 2022, and in total, 31 studies describing image processing methods were reviewed. The quantitative and qualitative analysis included 39 algorithms, which were selected as the most effective in this field. They were completed by 20 medical papers and define the needs and opportunities for high-frequency ultrasound application and CAD development.

Predicting F v /F m and evaluating cotton drought tolerance using hyperspectral and 1D-CNN

The chlorophyll fluorescence parameter F_v/F_m is significant in abiotic plant stress. Current acquisition methods must deal with the dark adaptation of plants, which cannot achieve rapid, real-time, and high-throughput measurements. However, increased inputs on different genotypes based on hyperspectral model recognition verified its capabilities of handling large and variable samples. F_v/F_m is a drought tolerance index reflecting the best drought tolerant cotton genotype. Therefore, F_v/F_m hyperspectral prediction of different cotton varieties, and drought tolerance evaluation, are worth exploring. In this study, 80 cotton varieties were studied. The hyperspectral cotton data were obtained during the flowering, boll setting, and boll opening stages under normal and drought stress conditions. Next, One-dimensional convolutional neural networks (1D-CNN), Categorical Boosting (CatBoost), Light Gradient Boosting Machines (LightBGM), eXtreme Gradient Boosting (XGBoost), Decision Trees (DT), Random Forests (RF), Gradient elevation decision trees (GBDT), Adaptive Boosting (AdaBoost), Extra Trees (ET), and K-Nearest Neighbors (KNN) were modeled with F _v /F _m. The Savitzky-Golay + 1D-CNN model had the best robustness and accuracy (RMSE = 0.016, MAE = 0.009, MAPE = 0.011). In addition, the F _v /F _m prediction drought tolerance coefficient and the manually measured drought tolerance coefficient were similar. Therefore, cotton varieties with different drought tolerance degrees can be monitored using hyperspectral full band technology to establish a 1D-CNN model. This technique is non-destructive, fast and accurate in assessing the drought status of cotton, which promotes smart-scale agriculture.

Automatic Segmentation of Laser-Induced Injury OCT Images Based on a Deep Neural Network Model

Optical coherence tomography (OCT) has considerable application potential in noninvasive diagnosis and disease monitoring. Skin diseases, such as basal cell carcinoma (BCC), are destructive; hence, quantitative segmentation of the skin is very important for early diagnosis and treatment. Deep neural networks have been widely used in the boundary recognition and segmentation of diseased areas in medical images. Research on OCT skin segmentation and laser-induced skin damage segmentation based on deep neural networks is still in its infancy. Here, a segmentation and quantitative analysis pipeline of laser skin injury and skin stratification based on a deep neural network model is proposed. Based on the stratification of mouse skins, a laser injury model of mouse skins induced by lasers was constructed, and the multilayer structure and injury areas were accurately segmented by using a deep neural network method. First, the intact area of mouse skin and the damaged areas of different laser radiation doses are collected by the OCT system, and then the labels are manually labeled by experienced histologists. A variety of deep neural network models are used to realize the segmentation of skin layers and damaged areas on the skin dataset. In particular, the U-Net model based on a dual attention mechanism is used to realize the segmentation of the laser-damage structure, and the results are compared and analyzed. The segmentation results showed that the Dice coefficient of the mouse dermis layer and injury area reached more than 0.90, and the Dice coefficient of the fat layer and muscle layer reached more than 0.80. In the evaluation results, the average surface distance (ASSD) and Hausdorff distance (HD) indicated that the segmentation results are excellent, with a high overlap rate with the manually labeled area and a short edge distance. The results of this study have important application value for the quantitative analysis of laser-induced skin injury and the exploration of laser biological effects and have potential application value for the early noninvasive detection of diseases and the monitoring of postoperative recovery in the future.

Semisupervised representative learning for measuring epidermal thickness in human subjects in optical coherence tomography by leveraging datasets from rodent models

SIGNIFICANCE

Morphological changes in the epidermis layer are critical for the diagnosis and assessment of various skin diseases. Due to its noninvasiveness, optical coherence tomography (OCT) is a good candidate for observing microstructural changes in skin. Convolutional neural network (CNN) has been successfully used for automated segmentation of the skin layers of OCT images to provide an objective evaluation of skin disorders. Such method is reliable, provided that a large amount of labeled data is available, which is very time-consuming and tedious. The scarcity of patient data also puts another layer of difficulty to make the model more generalizable.

AIM

We developed a semisupervised representation learning method to provide data augmentations.

APPROACH

We used rodent models to train neural networks for accurate segmentation of clinical data.

RESULT

The learning quality is maintained with only one OCT labeled image per volume that is acquired from patients. Data augmentation introduces a semantically meaningful variance, allowing for better generalization. Our experiments demonstrate the proposed method can achieve accurate segmentation and thickness measurement of the epidermis.

CONCLUSION

This is the first report of semisupervised representative learning applied to OCT images from clinical data by making full use of the data acquired from rodent models. The proposed method promises to aid in the clinical assessment and treatment planning of skin diseases.

High-Frequency Ultrasound Dataset for Deep Learning-Based Image Quality Assessment

This study aims at high-frequency ultrasound image quality assessment for computer-aided diagnosis of skin. In recent decades, high-frequency ultrasound imaging opened up new opportunities in dermatology, utilizing the most recent deep learning-based algorithms for automated image analysis. An individual dermatological examination contains either a single image, a couple of pictures, or an image series acquired during the probe movement. The estimated skin parameters might depend on the probe position, orientation, or acquisition setup. Consequently, the more images analyzed, the more precise the obtained measurements. Therefore, for the automated measurements, the best choice is to acquire the image series and then analyze its parameters statistically. However, besides the correctly received images, the resulting series contains plenty of non-informative data: Images with different artifacts, noise, or the images acquired for the time stamp when the ultrasound probe has no contact with the patient skin. All of them influence further analysis, leading to misclassification or incorrect image segmentation. Therefore, an automated image selection step is crucial. To meet this need, we collected and shared 17,425 high-frequency images of the facial skin from 516 measurements of 44 patients. Two experts annotated each image as correct or not. The proposed framework utilizes a deep convolutional neural network followed by a fuzzy reasoning system to assess the acquired data’s quality automatically. Different approaches to binary and multi-class image analysis, based on the VGG-16 model, were developed and compared. The best classification results reach 91.7% accuracy for the first, and 82.3% for the second analysis, respectively.

2021 international consensus statement on optical coherence tomography for basal cell carcinoma: image characteristics, terminology and educational needs

BACKGROUND

Despite the widespread use of optical coherence tomography (OCT) for imaging of keratinocyte carcinoma, we lack an expert consensus on the characteristic OCT features of basal cell carcinoma (BCC), an internationally vetted set of OCT terms to describe various BCC subtypes, and an educational needs assessment.

OBJECTIVES

To identify relevant BCC features in OCT images, propose terminology based on inputs from an expert panel and identify content for a BCC-specific curriculum for OCT trainees.

METHODS

Over three rounds, we conducted a Delphi consensus study on BCC features and terminology between March and September 2020. In the first round, experts were asked to propose BCC subtypes discriminable by OCT, provide OCT image features for each proposed BCC subtypes and suggest content for a BCC-specific OCT training curriculum. If agreement on a BCC-OCT feature exceeded 67%, the feature was accepted and included in a final review. In the second round, experts had to re-evaluate features with less than 67% agreement and rank the ten most relevant BCC OCT image features for superficial BCC, nodular BCC and infiltrative and morpheaphorm BCC subtypes. In the final round, experts received the OCT-BCC consensus list for a final review, comments and confirmation.

RESULTS

The Delphi included six key opinion leaders and 22 experts. Consensus was found on terminology for three OCT BCC image features: (i) hyporeflective areas, (ii) hyperreflective areas and (iii) ovoid structures. Further, the participants ranked the ten most relevant image features for nodular, superficial, infiltrative and morpheaform BCC. The target group and the key components for a curriculum for OCT imaging of BCC have been defined.

CONCLUSION

We have established a set of OCT image features for BCC and preferred terminology. A comprehensive curriculum based on the expert suggestions will help implement OCT imaging of BCC in clinical and research settings.

Deep-learning approach for automated thickness measurement of epithelial tissue and scab using optical coherence tomography

SIGNIFICANCE

In order to elucidate therapeutic treatment to accelerate wound healing, it is crucial to understand the process underlying skin wound healing, especially re-epithelialization. Epidermis and scab detection is of importance in the wound healing process as their thickness is a vital indicator to judge whether the re-epithelialization process is normal or not. Since optical coherence tomography (OCT) is a real-time and non-invasive imaging technique that can perform a cross-sectional evaluation of tissue microstructure, it is an ideal imaging modality to monitor the thickness change of epidermal and scab tissues during wound healing processes in micron-level resolution. Traditional segmentation on epidermal and scab regions was performed manually, which is time-consuming and impractical in real time.

AIM

We aim to develop a deep-learning-based skin layer segmentation method for automated quantitative assessment of the thickness of in vivo epidermis and scab tissues during a time course of healing within a rodent model.

APPROACH

Five convolution neural networks were trained using manually labeled epidermis and scab regions segmentation from 1000 OCT B-scan images (assisted by its corresponding angiographic information). The segmentation performance of five segmentation architectures was compared qualitatively and quantitatively for validation set.

RESULTS

Our results show higher accuracy and higher speed of the calculated thickness compared with human experts. The U-Net architecture represents a better performance than other deep neural network architectures with 0.894 at F1-score, 0.875 at mean intersection over union, 0.933 at Dice similarity coefficient, and 18.28  μm at an average symmetric surface distance. Furthermore, our algorithm is able to provide abundant quantitative parameters of the wound based on its corresponding thickness maps in different healing phases. Among them, normalized epidermal thickness is recommended as an essential hallmark to describe the re-epithelialization process of the rodent model.

CONCLUSIONS

The automatic segmentation and thickness measurements within different phases of wound healing data demonstrates that our pipeline provides a robust, quantitative, and accurate method for serving as a standard model for further research into effect of external pharmacological and physical factors.

The Importance of Mimicking Dermal-Epidermal Junction for Skin Tissue Engineering: A Review

There is a distinct boundary between the dermis and epidermis in the human skin called the basement membrane, a dense collagen network that creates undulations of the dermal-epidermal junction (DEJ). The DEJ plays multiple roles in skin homeostasis and function, namely, enhancing the adhesion and physical interlock of the layers, creating niches for epidermal stem cells, regulating the cellular microenvironment, and providing a physical boundary layer between fibroblasts and keratinocytes. However, the primary role of the DEJ has been determined as skin integrity; there are still aspects of it that are poorly investigated. Tissue engineering (TE) has evolved promising skin regeneration strategies and already developed TE scaffolds for clinical use. However, the currently available skin TE equivalents neglect to replicate the DEJ anatomical structures. The emergent ability to produce increasingly complex scaffolds for skin TE will enable the development of closer physical and physiological mimics to natural skin; it also allows researchers to study the DEJ effect on cell function. Few studies have created patterned substrates that could mimic the human DEJ to explore their significance. Here, we first review the DEJ roles and then critically discuss the TE strategies to create the DEJ undulating structure and their effects. New approaches in this field could be instrumental for improving bioengineered skin substitutes, creating 3D engineered skin, identifying pathological mechanisms, and producing and screening drugs.

Deep Learning-Based High-Frequency Ultrasound Skin Image Classification with Multicriteria Model Evaluation

This study presents the first application of convolutional neural networks to high-frequency ultrasound skin image classification. This type of imaging opens up new opportunities in dermatology, showing inflammatory diseases such as atopic dermatitis, psoriasis, or skin lesions. We collected a database of 631 images with healthy skin and different skin pathologies to train and assess all stages of the methodology. The proposed framework starts with the segmentation of the epidermal layer using a DeepLab v3+ model with a pre-trained Xception backbone. We employ transfer learning to train the segmentation model for two purposes: to extract the region of interest for classification and to prepare the skin layer map for classification confidence estimation. For classification, we train five models in different input data modes and data augmentation setups. We also introduce a classification confidence level to evaluate the deep model’s reliability. The measure combines our skin layer map with the heatmap produced by the Grad-CAM technique designed to indicate image regions used by the deep model to make a classification decision. Moreover, we propose a multicriteria model evaluation measure to select the optimal model in terms of classification accuracy, confidence, and test dataset size. The experiments described in the paper show that the DenseNet-201 model fed with the extracted region of interest produces the most reliable and accurate results.

Real-time deep learning assisted skin layer delineation in dermal optical coherence tomography

We present deep learning assisted optical coherence tomography (OCT) imaging for quantitative tissue characterization and differentiation in dermatology. We utilize a manually scanned single fiber OCT (sfOCT) instrument to acquire OCT images from the skin. The focus of this study is to train a U-Net for automatic skin layer delineation. We demonstrate that U-Net allows quantitative assessment of epidermal thickness automatically. U-Net segmentation achieves high accuracy for epidermal thickness estimation for normal skin and leads to a clear differentiation between normal skin and skin lesions. Our results suggest that a single fiber OCT instrument with AI assisted skin delineation capability has the potential to become a cost-effective tool in clinical dermatology, for diagnosis and tumor margin detection.

Deep learning approach to skin layers segmentation in inflammatory dermatoses

Monitoring skin layers with medical imaging is critical to diagnosing and treating patients with chronic inflammatory skin diseases. The high-frequency ultrasound (HFUS) makes it possible to monitor skin condition in different dermatoses. Accurate and reliable segmentation of skin layers in patients with atopic dermatitis or psoriasis enables the assessment of the treatment effect by the layer thickness measurements. The epidermis and the subepidermal low echogenic band (SLEB) are the most important for further diagnosis since their appearance is an indicator of different skin problems. In medical practice, the analysis, including segmentation, is usually performed manually by the physician with all drawbacks of such an approach, e.g., extensive time consumption and lack of repeatability. Recently, HFUS becomes common in dermatological practice, yet it is barely supported by the development of automated analysis tools. To meet the need for skin layer segmentation and measurement, we developed an automated segmentation method of both epidermis and SLEB layers. It consists of a fuzzy c-means clustering-based preprocessing step followed by a U-shaped convolutional neural network. The network employs batch normalization layers adjusting and scaling the activation to make the segmentation more robust. The obtained segmentation results are verified and compared to the current state-of-the-art methods addressing the skin layer segmentation. The obtained Dice coefficient equal to 0.87 and 0.83 for the epidermis and SLEB, respectively, proves the developed framework's efficiency, outperforming the other approaches.

Application of OCT-Derived Attenuation Coefficient in Acute Burn-Damaged Skin

BACKGROUND AND OBJECTIVES

There remains a need to objectively monitor burn wound healing within a clinical setting, and optical coherence tomography (OCT) is proving itself one of the ideal modalities for just such a use. The aim of this study is to utilize the noninvasive and multipurpose capabilities of OCT, along with its cellular-level resolution, to demonstrate the application of optical attenuation coefficient (OAC), as derived from OCT data, to facilitate the automatic digital segmentation of the epidermis from scan images and to work as an objective indicator for burn wound healing assessment.

STUDY DESIGN/MATERIALS AND METHODS

A simple, yet efficient, method was used to estimate OAC from OCT images taken over multiple time points following acute burn injury. This method enhanced dermal-epidermal junction (DEJ) contrast, which facilitated the automatic segmentation of the epidermis for subsequent thickness measurements. In addition, we also measured and compared the average OAC of the dermis within said burns for correlative purposes.

RESULTS

Compared with unaltered OCT maps, enhanced DEJ contrast was shown in OAC maps, both from single A-lines and completed B-frames. En face epidermal thickness and dermal OAC maps both demonstrated significant changes between imaging sessions following burn injury, such as a loss of epidermal texture and decreased OAC. Quantitative analysis also showed that OAC of acute burned skin decreased below that of healthy skin following injury.

CONCLUSIONS

Our study has demonstrated that the OAC estimated from OCT data can be used to enhance imaging contrast to facilitate the automatic segmentation of the epidermal layer, as well as help elucidate our understanding of the pathological changes that occur in human skin when exposed to acute burn injury, which could serve as an objective indicator of skin injury and healing.

Quantitative characterization of human breast tissue based on deep learning segmentation of 3D optical coherence tomography images

In this study, we performed dual-modality optical coherence tomography (OCT) characterization (volumetric OCT imaging and quantitative optical coherence elastography) on human breast tissue specimens. We trained and validated a U-Net for automatic image segmentation. Our results demonstrated that U-Net segmentation can be used to assist clinical diagnosis for breast cancer, and is a powerful enabling tool to advance our understanding of the characteristics for breast tissue. Based on the results obtained from U-Net segmentation of 3D OCT images, we demonstrated significant morphological heterogeneity in small breast specimens acquired through diagnostic biopsy. We also found that breast specimens affected by different pathologies had different structural characteristics. By correlating U-Net analysis of structural OCT images with mechanical measurement provided by quantitative optical coherence elastography, we showed that the change of mechanical properties in breast tissue is not directly due to the change in the amount of dense or porous tissue.

Deep convolutional neural networks: Outperforming established algorithms in the evaluation of industrial optical coherence tomography (OCT) images of pharmaceutical coatings

This paper presents a novel evaluation approach for optical coherence tomography (OCT) image analysis of pharmaceutical solid dosage forms based on deep convolutional neural networks (CNNs). As a proof of concept, CNNs were applied to image data from both, in- and at-line OCT implementations, monitoring film-coated tablets as well as single- and multi-layered pellets. CNN results were compared against results from established algorithms based on ellipse-fitting, as well as to human-annotated ground truth data. Performance benchmarks used include, efficiency (computation speed), sensitivity (number of detections from a defined test set) and accuracy (deviation from the reference method). The results were validated by comparing the output of several algorithms to data manually annotated by human experts and microscopy images of cross-sectional cuts of the same dosage forms as a reference method. In order to guarantee comparability for all results, the algorithms were executed on the same hardware. Since modern OCT systems must operate under real-time conditions in order to be implemented in-line into manufacturing lines, the necessary steps are discussed on how to achieve this goal without sacrificing the algorithmic performance and how to tailor a deep CNN to cope with the high amount of image noise and alterations in object appearance. The developed deep learning approach outperforms static algorithms currently available in pharma applications with respect to performance benchmarks, and represents the next level in real time evaluation of challenging industrial OCT image data.